The year 2007 marked three memorable events in climate science: Release of the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC AR4); a decade of drought in the American West and the arrival of severe drought in the American Southeast; and the disappearance of nearly half of the polar sea-ice floating over the Arctic Ocean. The IPCC report (a three-volume, three-thousand page synthesis of current scientific knowledge written for policymakers) and the American droughts merely hardened my conviction that anthropogenic climate warming is real and just getting going — a view shared, in the case of the IPCC, a few weeks ago by the Nobel Foundation. The sea-ice collapse, however, changed my mind that it will be decades before we see the real impacts of the warming. I now believe they will happen much sooner.
Let's put the 2007 sea-ice year into context. In the 1970's, when NASA first began mapping sea ice from microwave satellites, its annual minimum extent (in September, at summer's end) hovered close to 8 million square kilometers, about the area of the conterminous United States minus Ohio. In September 2007 it dropped abruptly to 4.3 million square kilometers, the area of the conterminous United State minus Ohio and all the other twenty-four states east of the Mississippi, as well as North Dakota, Minnesota, Missouri, Arkansas, Louisiana, and Iowa. Canada's Northwest Passage was freed of ice for the first time in human memory. From Bering Strait where the U.S. and Russia brush lips, open blue water stretched almost to the North Pole.
What makes the 2007 sea-ice collapse so unnerving is that it happened too soon. The ensemble averages of our most sophisticated climate model predictions, put forth in the IPCC AR4 report and various other model intercomparison studies, don't predict a downwards lurch of that magnitude for another fifty years. Even the aggressive models -the National Center for Atmospheric Research (NCAR) CCSM3 and the Centre National de Recherches Meteorologiques (CNRM) CM3 simulations, for example — must whittle ice until 2035 or later before the 2007 conditions can be replicated. Put simply, the models are too slow to match reality. Geophysicists, accustomed to non-linearities and hard to impress after a decade of 'unprecedented' events, are stunned by the totter: Apparently, the climate system can move even faster than we thought. This has decidedly recalibrated scientist's attitudes — including my own — to the possibility that even the direst IPCC scenario predictions for the end of this century — 10 to 24 inch higher global sea levels, for example — may be prudish.
What does all this say to us about the future? The first is that rapid climate change — a nonlinearity that occurs when a climate forcing reaches a threshold beyond which little additional forcing is needed to trigger a large impact — is a distinct threat not well captured in our current generation of computer models. This situation will doubtless improve — as the underlying physics of the 2007 ice event and others such as the American Southeast drought are dissected, understood, and codified — but in the meantime, policymakers must work from the IPCC blueprint which seems almost staid after the events of this summer and fall. The second is that it now seems probable that the northern hemisphere will lose its ice lid far sooner than we ever thought possible. Over the past three years experts have shifted from 2050, to 2035, to 2013 as plausible dates for an ice-free Arctic Ocean — estimates at first guided by models then revised by reality.
The broader significance of vanishing sea ice extends far beyond suffering polar bears, new shipping routes, or even development of vast Arctic energy reserves. It is absolutely unequivocal that the disappearance of summer sea ice — regardless of exactly which year it arrives — will profoundly alter the northern hemisphere climate, particularly through amplified winter warming of at least twice the global average rate. Its further impacts on the world's precipitation and pressure systems are under study but are likely significant. Effects both positive and negative, from reduced heating oil consumption to outbreaks of fire and disease, will propagate far southward into the United States, Canada, Russia and Scandinavia. Scientists have expected such things in eventuality — but in 2007 we learned they may already be upon us.